WO2016153479A1 - Balayage facial de flux vidéo - Google Patents

Balayage facial de flux vidéo Download PDF

Info

Publication number
WO2016153479A1
WO2016153479A1 PCT/US2015/022087 US2015022087W WO2016153479A1 WO 2016153479 A1 WO2016153479 A1 WO 2016153479A1 US 2015022087 W US2015022087 W US 2015022087W WO 2016153479 A1 WO2016153479 A1 WO 2016153479A1
Authority
WO
WIPO (PCT)
Prior art keywords
human
faces
camera
human object
unit
Prior art date
Application number
PCT/US2015/022087
Other languages
English (en)
Inventor
Qian Lin
Original Assignee
Longsand Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Longsand Limited filed Critical Longsand Limited
Priority to EP15886637.6A priority Critical patent/EP3275174A1/fr
Priority to PCT/US2015/022087 priority patent/WO2016153479A1/fr
Priority to US15/560,275 priority patent/US10719717B2/en
Publication of WO2016153479A1 publication Critical patent/WO2016153479A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G06T7/248—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments involving reference images or patches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/292—Multi-camera tracking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/161—Detection; Localisation; Normalisation
    • G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172—Classification, e.g. identification
    • G06V40/173—Classification, e.g. identification face re-identification, e.g. recognising unknown faces across different face tracks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/61—Control of cameras or camera modules based on recognised objects
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/61—Control of cameras or camera modules based on recognised objects
    • H04N23/611—Control of cameras or camera modules based on recognised objects where the recognised objects include parts of the human body
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/66—Remote control of cameras or camera parts, e.g. by remote control devices
    • H04N23/661—Transmitting camera control signals through networks, e.g. control via the Internet
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20076—Probabilistic image processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person
    • G06T2207/30201—Face
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30232—Surveillance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30241—Trajectory
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/44—Event detection
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/695—Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects

Definitions

  • Equipment such as security cameras, is used to provide surveillance video, especially in public areas. Providing more effective methods to identify people in video surveillance video from security cameras is becoming increasingly important.
  • FIG. 1 is an example block diagram of a system to scan a face of a video feed
  • FIG. 2 is another example block diagram of a system to scan a face of a video feed
  • FIG. 3 is an example block diagram of a computing device including instructions for scanning a face of a video feed
  • FIG. 4 is an example flowchart of a method for scanning a face of a video feed.
  • PTZ Pan, Tilt, and Zoom
  • Traditionally PTZ cameras have been controlled either manually or via simple logic such as detecting faces and zooming in to faces or static speed thresholds for human walking. Further, such methods may be difficult to scale for a large number of cameras.
  • Examples may provide a system that controls PTZ cameras by actively analyzing the crowd behavior in the video. Upon detecting unusual behaviors, examples may start automatic scanning of the faces to create a record of the people present at the event.
  • the system may include an analysis unit and an event unit.
  • the analysis unit may analyze crowd behavior in a video feed based on movement of a human object identified in the video feed.
  • the event unit may trigger an abnormal event if the movement of the human object deviates from the analyzed crowd behavior.
  • a camera may scan a face of the human object, if the abnormal event is triggered.
  • examples may describe an adaptive system to detect anomalies by analyzing the statistics of human body movement observed by cameras. Examples may not require manual steps for calibration.
  • examples may Integrate with a PTZ camera's pan, tilt, and zoom function with human body detection and tracking result allows the scanning of faces at high resolution once anomalies are detected.
  • the face images as well as the associated appearance time and location data may provide a relatively fast way to index and search videos captured by a large number of cameras at an event location as well as adjacent areas.
  • the faces may also be used to match with existing databases to obtain the identities of the face images.
  • FIG. 1 is an example block diagram of a system 100 to scan a face of a video feed.
  • the system 100 may include a microprocessor, a controller, a memory module or device, a notebook computer, a desktop computer, an all-in-one system, a server, a network device, a wireless device, a storage device, a surveillance device such as a camera or sensor, or any other type of device capable of controlling a camera.
  • the system 200 is shown to include an analysis unit 1 10 and an event unit 120.
  • the analysis and event units 1 10 and 120 may include, for example, a hardware device including electronic circuitry for implementing the functionality described below, such as control logic and/or memory, in addition or as an alternative, the analysis and event units 1 10 and 120 may be implemented as a series of instructions encoded on a machine-readable storage medium and executable by a processor.
  • the analysis unit 1 10 may analyze crowd behavior 1 12 in a video feed based on movement 1 14 of a human object identified in the video feed.
  • the event unit 120 may trigger an abnormal event 122 if the movement 1 14 of the human object deviates from the analyzed crowd behavior 1 12.
  • a camera may scan a face of the human object, if the abnormal event 122 is triggered.
  • FIG. 2 is another example block diagram of a system 200 to scan a face of a video feed.
  • the system 200 may include a microprocessor, a controller, a memory module or device, a notebook computer, a desktop computer, an all-in-one system, a server, a network device, a wireless device, a storage device, a surveillance device such as a camera or sensor, or any other type of device capable of controlling a camera.
  • the system 200 of FIG. 2 may include at least the functionality and/or hardware of the system 100 of FIG. 1.
  • the system 200 is shown to include an analysis unit 210 and an event unit 220.
  • the analysis unit 210 of the system 200 of FIG. 2 may include at least the functionality and/or hardware of the analysis unit 1 10 of the system 100 of FIG. 1 and the event unit 220 of the system 200 of FIG. 2 may include at least the functionality and/or hardware of the event unit 120 of the system 100 of FIG. 1.
  • the system 200 is further shown to include a detection unit 230, a tracking unit 240 and a scanning unit 250.
  • the detection, tracking and scanning units 230, 240 and 250 may include, for example, a hardware device including electronic circuitry for implementing the functionality described below, such as control logic and/or memory.
  • the detection, tracking and scanning units 230, 240 and 250 may be implemented as a series of instructions encoded on a machine-readable storage medium and executable by a processor.
  • the system 200 is also shown to interface with a plurality of cameras 260-1 to 260-n, where n is a natural number.
  • the cameras 280 may be optical devices that record images and may be used for electronic motion picture acquisition.
  • the cameras 280 may be pan-ti!t-zoom (PTZ) cameras that are capable of remote directional and zoom control.
  • PTZ pan-ti!t-zoom
  • the system 200 may receive one or more video feeds 282 from the cameras 260.
  • the detection unit 230 may detect one or more human objects in the video feeds 262.
  • the human object may be detected by analyzing the plurality of pixels of a video frame of the video feed 262 using a Histogram of Oriented Gradient (HOG) 232 and/or a cascading classifier 234.
  • HOG Histogram of Oriented Gradient
  • the tracking unit 240 may track a position of the human object over a time interval to determine a trajectory.
  • the tracking unit 240 may track a center of the human object to determine trajectory coordinates 242.
  • the analysis unit 210 may calculate a velocity 212 of the center of the human object.
  • the velocity 212 may include at least one of a speed and a direction of the human object. For example, if the human object comprises a plurality of pixels, a center-most pixel of the plurality of the pixels may be used to track the velocity 212 of the human object.
  • the tracking unit 240 may track a plurality of the human objects in the video feed 262.
  • the analysis unit 210 may determine the crowd behavior 1 12 based on a plurality of velocities 212 of the plurality of human objects.
  • the analysis unit 210 may determine a probability distribution 214 based on the plurality of velocities 212, such as a normal distribution.
  • the event unit 222 may trigger the abnormal event 122 if an observed velocity 212 is greater than a velocity threshold 222 number of standard deviations from a mean of the probability distribution 214, For example, the velocity threshold 222 may be three. Thus, a resulting velocity 212 that is greater than three standard deviations would trigger the abnormal event 122.
  • a person may panic and start running down a hallway, due to a terrorist act, accident, or the like. This person may be recognized as a human object and their velocity may be measured and compared to the distribution 214 to determine a number of standard deviations.
  • the analysis unit 210 may aggregate the plurality of velocities 212 to determine a pattern 216 for the plurality of pixels positions 264 across the video frame during a normal state.
  • the normal state may reflect a time period during which no abnormal events are occurring.
  • the event unit 220 may trigger the abnormal event 122 if the crowd behavior is significantly different from the pattern 216 at more than a pattern threshold number 228 (e.g. 10% or 25% of pixel positions 264) of the plurality of pixel positions 264.
  • the analysis unit 210 may observe a pattern 216 where human objects are pacing back and forth, such as near a train stop, if one or more humans started running, this pattern 216 may be broken and replaced by randomness, or a different pattern, such as a radial pattern or human objects only flowing in one direction. This change may be reflected at more than the pattern threshold 228 number of pixel positions 264.
  • the analysis unit 210 may dynamically calculate the average velocity vector 224 for at least one of the pixel positions 264 between adjacent video frames during the normal state.
  • the event unit 220 may trigger the abnormal event 122 if an observed velocity of the human object at the at least one pixel position 264 exceeds the average velocity 224 of the at ieast one pixel position 264 by a pixel threshold 226 for the at least one pixel position 264.
  • the average velocity 224 for the pixel positions 264 along the major walkway may be based on a greater number of observed human objects and/or different human object behavior.
  • the average velocity 224 for the pixel positions 264 along the major walkway may be different than that of the minor walkway.
  • the pixel threshold 226 for the pixel positions 264 along the major walkway may be different than that of the minor walkway.
  • the pixel thresholds 226 for the pixel positions 264 along the major walkway may span a greater range than that of the minor walkway.
  • a same observed velocity that may not trigger the abnormal event 122 for pixel positions 264 along the major walkway could still trigger the abnormal event 122 for pixel positions 264 along the minor walkway.
  • pixel positions 264 where there is no human traffic may have even lower average velocities 224 and/or pixel threshold 226.
  • the scanning unit 250 may direct the camera 260 to scan a plurality of faces of the plurality of human objects in the video frame, if the abnormal event 122 is triggered.
  • the camera 260 may pan, tile and/or zoom to scan the plurality of faces.
  • the scanning unit 250 may direct the camera 260 to scan the plurality of faces sequentially.
  • the scanning unit 250 may scan at Ieast one of the plurality of faces at a higher resolution in response to the camera 260 zooming in.
  • the scanning unit 250 may direct the camera 260 to assign a higher priority 252 to scanning the faces of the human objects that are closer to a boundary of the video frame, in order to scan these faces before the human objects leave the field of view of the camera 260.
  • the scanning unit 250 may direct the camera 260 to assign higher priority 252 to scanning the faces of the human objects that are less than a threshold 254 number of degrees between a line defined by the camera and the face, and the horizon, such as fifteen degrees. Also, the scanning unit 250 may direct the camera 260 to assign higher priority 252 to scanning the faces of the human objects that are further away from the camera 260. This is because a closer distance at a steeper angle may result in lower visibility of the human object's face.
  • the scanning unit 250 may direct the camera 260 to assign a higher priority 252 to scanning a frontal-view of the faces of the human objects, in order to improve facial recognition.
  • the scanning unit 250 may create an index 256 of scanned faces.
  • the index 256 may include at least one of associated time and location data.
  • the index 256 may be used to match the scanned faces with the faces of a database (not shown) via a face recognition system.
  • the faces may be scanned by the plurality of cameras 260-1 to 260-n in real time.
  • the scanning unit 250 may instruct the camera 260 to focus upon a center of a first human object of the plurality of human objects. Then, the scanning unit 250 may instruct the camera 260 to zoom to at least one of a height of the first human object and a maximum zoom of the camera 260 (if the zoom is limited). Next, the scanning unit 250 may instruct the camera 260 to track the first human object while a face of the first human object is scanned. Lastly, the scanning unit 250 may instruct the camera 280 to repeat the above focusing, zooming and tracking for a second human object of the plurality of human objects, after the face of the first human object is scanned.
  • a field of vision of at least two of the plurality of cameras 260-1 to 260-n may overlap.
  • the priority 252 for scanning at least one of the human objects for one of the plurality of cameras 260-1 to 260-n may be based on whether the face of the at least one human object has been scanned by another of the plurality of cameras 260-1 to 260-n.
  • examples may provide a system for monitoring a crowd using a video surveillance camera that has pan, tilt, and zoom control.
  • Examples may analyze crowd behavior from the surveillance video.
  • examples may enact pan, tile, and zoom controls to scan faces to create an index of people present at the event. The faces may then be matched with enrolled faces in databases using face recognition system to obtain the identity of the faces.
  • FIG. 3 is an example block diagram of a computing device 300 including instructions for scanning a face of a video feed.
  • the computing device 300 includes a processor 310 and a machine-readable storage medium 320.
  • the machine-readable storage medium 320 further includes instructions 322, 324, 326 and 328 for scanning a face of a video feed.
  • the computing device 300 may be included in or part of, for example, a microprocessor, a controller, a memory module or device, a notebook computer, a desktop computer, an all-in-one system, a server, a network device, a wireless device, a storage device, a surveillance device such as a camera, or any other type of device capable of executing the instructions 322, 324, 328 and 328.
  • the computing device 300 may include or be connected to additional components such as memories, controllers, etc.
  • the processor 310 may be, at least one central processing unit (CPU), at least one semiconductor-based microprocessor, at least one graphics processing unit (GPU), a microcontroller, special purpose logic hardware controlled by microcode or other hardware devices suitable for retrieval and execution of instructions stored in the machine-readable storage medium 320, or combinations thereof.
  • the processor 310 may fetch, decode, and execute instructions 322, 324, 326 and 328 to implement scanning the face of the video feed.
  • the processor 310 may include at least one integrated circuit (IC), other control logic, other electronic circuits, or combinations thereof that include a number of electronic components for performing the functionality of instructions 322, 324, 326 and 328.
  • IC integrated circuit
  • the machine-readable storage medium 320 may be any electronic, magnetic, optical, or other physical storage device that contains or stores executable instructions.
  • the machine-readable storage medium 320 may be, for example, Random Access Memory (RAM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a storage drive, a Compact Disc Read Only Memory (CD-ROM), and the like.
  • RAM Random Access Memory
  • EEPROM Electrically Erasable Programmable Read-Only Memory
  • CD-ROM Compact Disc Read Only Memory
  • the machine- readable storage medium 320 can be non-transitory.
  • machine-readable storage medium 320 may be encoded with a series of executable instructions for scanning the face of the video feed.
  • the instructions 322, 324, 328 and 328 when executed by a processor (e.g., via one processing element or multiple processing elements of the processor) can cause the processor to perform processes, such as, the process of FIG. 4.
  • the identify instructions 322 may be executed by the processor 310 to identify trajectories of a plurality of human objects across a plurality of pixel positions for a plurality of frames of a video feed.
  • the calculate average instructions 324 may be executed by the processor 310 to calculate an average velocity vector for the plurality of pixel positions based on the trajectories.
  • the calculate threshold instructions 326 may be executed by the processor 310 to calculate a plurality of pixel thresholds for each of the plurality of pixel positions based on the average velocity vectors.
  • the scan instructions 328 may be executed by the processor 310 to scan faces of the plurality of human objects if a measured velocity vector for at least one of the plurality of human objects exceeds the pixel threshold for the associated pixel position.
  • the faces may be scanned via a plurality of the PTZ cameras (not shown) in real time.
  • a field of vision of at least two of the plurality of PTZ cameras may overlap.
  • a priority for scanning at least one of the human objects for one of the plurality of PTZ cameras may be based on whether the face of the at least one human object has been scanned by another of the plurality of PTZ cameras.
  • FIG. 4 is an example flowchart of a method 400 for scanning a face of a video feed. Although execution of the method 400 is described below with reference to the system 200, other suitable components for execution of the method 400 can be utilized, such as the system 100.
  • the components for executing the method 400 may be spread among multiple devices (e.g., a processing device in communication with input and output devices). In certain scenarios, multiple devices acting in coordination can be considered a single device to perform the method 400.
  • the method 400 may be implemented in the form of executable instructions stored on a machine-readable storage medium, such as storage medium 320, and/or in the form of electronic circuitry.
  • the system 200 calculates velocities 212 of a plurality of human objects of a video feed 262 across a plurality of pixels positions 264 over a time interval. Then, at block 420, the system 200 determines crowd behavior 1 12 during a normal state based on an average velocity 224 for at least one of the plurality of pixel positions 264. Next, at block 430, the system 200 compares an observed velocity of the human object at the at least one pixel position 264 to the average velocity 224 for the at least one pixel position 264.
  • the system 200 triggers a camera 260 to at least one of pan, tilt and zoom to scan faces of the plurality of human of objects of the video feed 262, if the observed velocity exceeds the average velocity 224 by a pixel threshold 226 of the at least one pixel position 264.
  • the triggering at block 440 may Include, for example, focusing at a center of a first human object of the plurality of human objects and zooming to a height of the first human object or a maximum zoom of the camera 260 (if the zoom is limited).
  • the triggering at block 440 may further include tracking the first human object while a face of the first human object is scanned and repeating the focusing, zooming and tracking for a second human object of the plurality of human objects, after the face of the first human object is scanned.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • Library & Information Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Image Analysis (AREA)
  • Studio Devices (AREA)

Abstract

Un comportement de foule peut être analysé dans un flux vidéo d'après le mouvement d'un objet humain identifié dans le flux vidéo. Un événement anormal peut être déclenché si le mouvement de l'objet humain s'écarte du comportement de foule analysé. Une caméra peut balayer un visage de l'objet humain si l'événement anormal est déclenché.
PCT/US2015/022087 2015-03-23 2015-03-23 Balayage facial de flux vidéo WO2016153479A1 (fr)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP15886637.6A EP3275174A1 (fr) 2015-03-23 2015-03-23 Balayage facial de flux vidéo
PCT/US2015/022087 WO2016153479A1 (fr) 2015-03-23 2015-03-23 Balayage facial de flux vidéo
US15/560,275 US10719717B2 (en) 2015-03-23 2015-03-23 Scan face of video feed

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/US2015/022087 WO2016153479A1 (fr) 2015-03-23 2015-03-23 Balayage facial de flux vidéo

Publications (1)

Publication Number Publication Date
WO2016153479A1 true WO2016153479A1 (fr) 2016-09-29

Family

ID=56978570

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2015/022087 WO2016153479A1 (fr) 2015-03-23 2015-03-23 Balayage facial de flux vidéo

Country Status (3)

Country Link
US (1) US10719717B2 (fr)
EP (1) EP3275174A1 (fr)
WO (1) WO2016153479A1 (fr)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106534784A (zh) * 2016-11-22 2017-03-22 苏州航天系统工程有限公司 一种用于视频分析数据结果集的采集分析存储统计系统
CN109241933A (zh) * 2018-09-21 2019-01-18 深圳市九洲电器有限公司 视频联动监控方法、监控服务器、视频联动监控系统
CN109376689A (zh) * 2018-11-20 2019-02-22 图普科技（广州）有限公司 人群分析方法及装置
CN113553468A (zh) * 2020-04-24 2021-10-26 杭州海康威视数字技术股份有限公司 录像索引生成方法以及录像回放检索方法
CN113572998A (zh) * 2021-09-22 2021-10-29 中科南京智能技术研究院 一种基于事件相机的数据集采集方法及系统
CN113724214A (zh) * 2021-08-23 2021-11-30 唯智医疗科技(佛山)有限公司 一种基于神经网络的图像处理方法及装置

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106875422B (zh) * 2017-02-06 2022-02-25 腾讯科技（上海）有限公司 人脸跟踪方法和装置
CN109376601B (zh) * 2018-09-21 2021-05-11 深圳市九洲电器有限公司 基于高速球的物体跟踪方法、监控服务器、视频监控系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080201116A1 (en) * 2007-02-16 2008-08-21 Matsushita Electric Industrial Co., Ltd. Surveillance system and methods
WO2014056537A1 (fr) * 2012-10-11 2014-04-17 Longsand Limited Utilisation d'un modèle probabiliste pour détecter un objet dans des données visuelles
US20140152821A1 (en) * 2012-11-30 2014-06-05 Li-Chen Fu Pedestrian detection systems and methods
US20140313330A1 (en) * 2013-04-19 2014-10-23 James Carey Video identification and analytical recognition system
US20140372348A1 (en) * 2011-12-15 2014-12-18 Northeastern University Real-time anomaly detection of crowd behavior using multi-sensor information

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6965645B2 (en) * 2001-09-25 2005-11-15 Microsoft Corporation Content-based characterization of video frame sequences
US20050104958A1 (en) 2003-11-13 2005-05-19 Geoffrey Egnal Active camera video-based surveillance systems and methods
GB0507869D0 (en) 2005-04-19 2005-05-25 Wqs Ltd Automated surveillance system
CN101918989B (zh) 2007-12-07 2013-02-13 常州环视高科电子科技有限公司 带有对象跟踪和检索的视频监控系统
US20090158309A1 (en) * 2007-12-12 2009-06-18 Hankyu Moon Method and system for media audience measurement and spatial extrapolation based on site, display, crowd, and viewership characterization
US8284258B1 (en) 2008-09-18 2012-10-09 Grandeye, Ltd. Unusual event detection in wide-angle video (based on moving object trajectories)
WO2013003860A1 (fr) * 2011-06-30 2013-01-03 Yale University Détection de sujets dans un environnement
JP5944781B2 (ja) * 2012-07-31 2016-07-05 株式会社デンソーアイティーラボラトリ 移動体認識システム、移動体認識プログラム、及び移動体認識方法
CN102970517B (zh) 2012-11-28 2015-08-19 四川长虹电器股份有限公司 基于异常情景识别的云台镜头自主控制方法
US9762865B2 (en) * 2013-03-15 2017-09-12 James Carey Video identification and analytical recognition system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080201116A1 (en) * 2007-02-16 2008-08-21 Matsushita Electric Industrial Co., Ltd. Surveillance system and methods
US20140372348A1 (en) * 2011-12-15 2014-12-18 Northeastern University Real-time anomaly detection of crowd behavior using multi-sensor information
WO2014056537A1 (fr) * 2012-10-11 2014-04-17 Longsand Limited Utilisation d'un modèle probabiliste pour détecter un objet dans des données visuelles
US20140152821A1 (en) * 2012-11-30 2014-06-05 Li-Chen Fu Pedestrian detection systems and methods
US20140313330A1 (en) * 2013-04-19 2014-10-23 James Carey Video identification and analytical recognition system

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106534784A (zh) * 2016-11-22 2017-03-22 苏州航天系统工程有限公司 一种用于视频分析数据结果集的采集分析存储统计系统
CN109241933A (zh) * 2018-09-21 2019-01-18 深圳市九洲电器有限公司 视频联动监控方法、监控服务器、视频联动监控系统
CN109376689A (zh) * 2018-11-20 2019-02-22 图普科技（广州）有限公司 人群分析方法及装置
CN113553468A (zh) * 2020-04-24 2021-10-26 杭州海康威视数字技术股份有限公司 录像索引生成方法以及录像回放检索方法
CN113553468B (zh) * 2020-04-24 2023-07-14 杭州海康威视数字技术股份有限公司 录像索引生成方法以及录像回放检索方法
CN113724214A (zh) * 2021-08-23 2021-11-30 唯智医疗科技(佛山)有限公司 一种基于神经网络的图像处理方法及装置
CN113724214B (zh) * 2021-08-23 2024-02-23 唯智医疗科技(佛山)有限公司 一种基于神经网络的图像处理方法及装置
CN113572998A (zh) * 2021-09-22 2021-10-29 中科南京智能技术研究院 一种基于事件相机的数据集采集方法及系统

Also Published As

Publication number Publication date
US20180075307A1 (en) 2018-03-15
US10719717B2 (en) 2020-07-21
EP3275174A1 (fr) 2018-01-31

Similar Documents

Publication Publication Date Title
US10719717B2 (en) Scan face of video feed
KR102195706B1 (ko) 침입 탐지방법 및 그 장치
US8754945B2 (en) Image capturing device and motion tracking method
JP5390322B2 (ja) 画像処理装置、及び画像処理方法
WO2014125882A1 (fr) Système de traitement d'informations, procédé de traitement d'informations et programme
KR101530255B1 (ko) 객체 자동 추적 장치가 구비된 cctv 시스템
KR101858396B1 (ko) 지능형 침입 탐지 시스템
JP7101805B2 (ja) ビデオ異常検出のためのシステム及び方法
CN108875507B (zh) 行人跟踪方法、设备、系统和计算机可读存储介质
US20190304272A1 (en) Video detection and alarm method and apparatus
CN113033521B (zh) 基于目标分析的周界动态预警方法与系统
US8923552B2 (en) Object detection apparatus and object detection method
WO2018064773A1 (fr) Système de surveillance vidéo et dispositif de dissuasion physique combinés
Juhana et al. Design and implementation of Smart Home Surveillance system
US20200293759A1 (en) Online face clustering
JP4610005B2 (ja) 画像処理による侵入物体検出装置および方法およびプログラム
KR100706871B1 (ko) 감시영상에서 얼굴의 진위 여부를 구별하는 방법
KR102142315B1 (ko) 영상 분석 기반의 금융자동화기기 보안 시스템 및 그 방법
US11164008B2 (en) Method and controller for controlling a video processing unit to facilitate detection of newcomers in a first environment
KR102113533B1 (ko) 영상 내의 이상 행동 객체를 추적하는 장치
JPH09252467A (ja) 移動体検出装置
KR102046591B1 (ko) 영상 감시 시스템 및 영상 감시방법
CN114764895A (zh) 异常行为检测装置和方法
KR101154350B1 (ko) 고해상도 영상의 객체 검지 및 분할 인식을 이용한 다중 영상처리시스템 및 방법
Chiperi et al. Data fusion in a multi agent system for person detection and tracking in an intelligent room

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15886637

Country of ref document: EP

Kind code of ref document: A1

REEP Request for entry into the european phase

Ref document number: 2015886637

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 15560275

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE